Effect of Electromagnetic Vibrations on Fe-Co-B-Si-Nb Bulk Metallic Glasses

Transcription

1 Materials Transactions, Vol. 48, No. 1 (2007) pp. 53 to 57 #2007 The Japan Institute of Metals Effect of Electromagnetic Vibrations on Fe-Co-B-Si-Nb Bulk Metallic Glasses Takuya Tamura*, Daisuke Kamikihara, Naoki Omura and Kenji Miwa Solidification Processing Group, Materials Research Institute for Sustainable Development, National Institute of Advanced Industrial Science and Technology (AIST), Nagoya , Japan It is known that cooling rate from the liquid state is an important factor for producing the bulk metallic glasses. However, almost no other factors such as electric and/or magnetic fields were investigated. The present authors have reported that a new method for producing Mg-Cu-Y bulk metallic glasses by using electromagnetic vibrations is effective in forming the metallic glass phase. Moreover, the present authors have reported that the glass-forming ability of Fe-Co-B-Si-Nb alloys also is enhanced with increasing the electromagnetic vibration force. Thus, this study aims to investigate effects of the electromagnetic vibrations on Fe-Co-B-Si-Nb bulk metallic glasses in order to investigate further. Half round lines which consist of fine crystal particles in a glassy phase were observed in the boundary part of the molybdenum electrode for the alloy with the electromagnetic vibrations. It was considered that the reactants of the molybdenum electrode which were solid at 1573 K were moved into the sample by the electromagnetic vibrations and caused the nuclei of crystal particles which composed the half round lines. If tungsten was used as the electrodes, there was no influence of electrodes in the center of the samples. When molybdenum or tungsten was used as the electrodes, the effect of the electromagnetic vibrations was found to be the same, namely the electromagnetic vibrations act on decreasing the number of crystal nuclei. [doi: /matertrans.48.53] (Received September 5, 2006; Accepted November 15, 2006; Published December 25, 2006) Keywords: bulk metallic glass, electromagnetic vibration, crystal growth, crystal nuclei, glass-forming ability 1. Introduction Different combinations of stationary and/or alternating electric and magnetic fields have been used for a wide range of purposes, including stirring, shaping, etc. 1) Among these combinations, it was reported that electromagnetic vibrations induced by the interaction of alternating electric and stationary magnetic fields can act as powerful vibrating forces in the melt and affect microstructural refinements in the usual crystalline alloys. 2 9) The simultaneous imposition of a stationary magnetic field with a magnetic flux density B and an alternating electric field with a current density J and a frequency f on a conducting liquid results in the induction of a vibrating electromagnetic body force with a density of F ¼ J B inside the liquid. This force, which has a frequency equal to that of the applied electric field, vibrates in a direction perpendicular to the plane of the two fields and puts particles constituting the conducting liquid into a vibrating motion. The present authors reported that a new method for producing Mg-Cu-Y bulk metallic glasses by using the electromagnetic vibrations is effective in forming the metallic glass phase, and disappearance or decrement of clusters by the electromagnetic vibrations applied to the liquid state is presumed to cause suppression of crystal nucleation. 10,11) Moreover, the present authors reported that it was found that the glass-forming ability of Fe-Co-B-Si-Nb alloys also increases with increasing the intensity of electromagnetic vibrations. The decrement of crystal particles with increasing the intensity of electromagnetic vibrations differed from that with increasing the cooling rate. The electromagnetic vibrations were found to act mainly on decreasing the number of crystal nuclei if one crystal particle is assumed to grow from one crystal nucleus. 12) However, effects of the electromagnetic vibrations on glass-forming ability are not fully investigated so far. Thus, the purpose of this study is to *Corresponding author, takuya-tamura@aist.go.jp investigate the effect of the electromagnetic vibrations on Fe- Co-B-Si-Nb bulk metallic glasses. 2. Experimental Procedures The (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloys with a diameter of 2 mm were provided by RIMCOF Japan. Alloy ingots were prepared by induction melting the mixtures of pure metals under an argon atmosphere. The cylindrical samples were produced by casting into a copper mold. It was reported that the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 bulk metallic glass rods were produced in a diameter of up to 3 mm by casting into a copper mold. 13,14) The (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy, which is cylindrical with a diameter of 2 mm and a length of 8 mm, was placed in an alumina tube of almost the same inner diameter and an outer diameter of 4 mm. The sample was held between two molybdenum or tungsten electrodes. Details of the experimental apparatus has been described in the previous paper. 11) Argon was passed through the inside of the stainless-steel cylinders to fix the sample in a superconductingmagnet. The sample was heated at a rate of 20 K/min to 1573 K by the heating furnace which was set around the sample placed at the center of the magnet. The molten sample was kept at this temperature for 2 minutes and then water was sprayed on the alumina tube. The flow rate of cooling water was fixed at 3 L/ min. The electromagnetic vibrations were applied by passing the alternating electric current through the sample at a preset magnetic field. The alternating electric current was fixed at 5 A. Applied time of the electromagnetic vibrations for the alloys is shown in Fig. 1. The electromagnetic vibrations were applied for 10 seconds before the onset of the water spray and also 10 seconds after that. The metallic glass or crystalline structures were examined by optical microscopy and HRTEM. For optical microscopy observations, 10% nitric acid-ethanol solution was used for etching the sample surface.

2 54 T. Tamura, D. Kamikihara, N. Omura and K. Miwa Fig. 1 Applied time of the electromagnetic vibrations. cooled under electromagnetic vibrations at a magnetic field of 10 T and an alternating electric current of 5 A, 5 khz. The HRTEM image corresponds to the optical micrograph and the XRD pattern for the alloy showed in the previous paper. 12) The diffraction pattern taken from the region of 400 nm in diameter consists of a halo ring. Only modulated contrast typical of a glassy single structure is observed. These data reveal that the alloy produced by the electromagnetic vibration process consists of a glassy phase. However, small crystal particles were observed in a glassy phase even if the alloy was produced by casting into a copper mold. Figure 3 shows a high-resolution transmission electron microscope image (HRTEM) of the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy produced by casting into a copper mold. The crystal particle with a diameter of about 3 mm was observed in a glassy phase. This crystal particle is considered to consist of Fe 3 B structure as determined by a diffraction pattern taken from the region of 800 nm in diameter. The crystal particle grew in the direction of outside. Moreover, the crystal nucleus is considered to be the center of this crystal particle because of considering the growing direction. This result proves the assumption, that one crystal particle grew from one crystal nucleus in (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy, suggested in the previous paper. 12) Fig. 2 HRTEM image of the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy cooled under electromagnetic vibrations at a magnetic field of 10 T and an alternating electric current of 5 A, 5 khz. The inset displays the selected-area electron diffraction pattern taken from the circular region of 400 nm in diameter. 3. Results and Discussion 3.1 The measurement by TEM Figure 2 shows a high-resolution transmission electron microscope image (HRTEM) and a selected-area electron diffraction pattern of the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy 3.2 The effects of the electrodes on glass-forming ability Figure 4 shows optical micrographs for the (Fe 0:6 - Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy cooled under electromagnetic vibrations. The alternating electric current and the magnetic field were 5 A, 5 khz and 10 T, respectively. The optical micrograph of boundary part of the molybdenum electrode is shown in Fig. 4(a). The optical micrograph of center part of the sample is shown in Fig. 4(b). The left edge part of the optical micrograph shown in Fig. 4(a) is the Mo electrode and another part is the sample. Half round lines which consist of crystal particles in a glassy phase were observed in the Fig. 3 HRTEM image of the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy produced by casting into a copper mold.

3 Effect of Electromagnetic Vibrations on Fe-Co-B-Si-Nb Bulk Metallic Glasses 55 Fig. 4 Optical micrographs for the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloy cooled under electromagnetic vibrations. The alternating electric current and the magnetic field were 5 A, 5 khz and 10 T, respectively. (a): optical micrograph of boundary part of the molybdenum electrode. (b): optical micrograph of center part of the sample. Fig. 5 SEM composition images corresponding to the optical micrograph shown in Fig. 4(a). (a): SEM image of magnified boundary part of the molybdenum electrode. (b): SEM image of crystal particles in the part of half round lines which consist of crystal particles. boundary part of the molybdenum electrode as shown in Fig. 4(a). However, half round lines which consist of crystal particles in a glassy phase were not observed in the center part of the sample shown in Fig. 4(b). Moreover, half round lines which consist of crystal particles didn t appear without the electromagnetic vibrations. Figure 5 shows SEM composition images corresponding to the optical micrograph shown in Fig. 4(a). The SEM image of magnified boundary part of the molybdenum electrode is shown in Fig. 5(a). The SEM image of crystal particles in the part of half round lines which consist of crystal particles is shown in Fig. 5(b). Whether it is metallic glass or crystal can be distinguished by the contrast of SEM image. Dark gray part is the crystalline phase, and light gray part is the glassy phase. This result is in agreement with that of the optical micrograph. Table 1 shows Mo content by SEM-EDS corresponding to the SEM images shown in Fig. 5. Boron Table 1 Mo content by SEM-EDS corresponding to the SEM images shown in Fig. 5. Boron was excluded from detection elements because boron cannot be detected exactly by SEM-EDS. (A) (B) (C) (D) (E) Mo (at%) was excluded from detection elements because boron cannot be detected exactly by SEM-EDS. Mo content at point (A) is 98%. Thus, this region is the Mo electrode. Points (C) and (D) are the region of the sample. Difference of points (C) and (D) is that point (D) is the metallic glass, but point (C) is the crystal region grown from the electrode. However, Mo content was not different. Mo content at point (B) is 18%. Thus, It is considered that point (B) is the reactants of the Mo electrode, and these reactants are solid at 1573 K. Mo content

4 56 T. Tamura, D. Kamikihara, N. Omura and K. Miwa Fig. 6 Optical micrographs of center part of the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloys. (a), (b): molybdenum was used as the electrodes. (c), (d): tungsten was used as the electrodes. The alternating electric current was fixed at 5 A, 5 khz. The magnetic fields were (a), (c) 0 T and (b), (d) 10 T. at point (E) which are in the center of crystal particles is 5%. A large number of small particles, which are in the center of crystal particles and have the same contrast as point (E) has, was observed in the part of half round lines which consist of crystal particles. These particles are considered to have the same Mo content as point (E) has because these SEM images are composition images. Thus, it is considered that the reactants of the Mo electrode at point (B) were moved into the sample by the electromagnetic vibrations and caused the nuclei of crystal particles which composed the half round lines. However, the electromagnetic vibration effect cannot be evaluated exactly if the samples contain Mo by the electromagnetic vibrations. If tungsten is used as the electrodes, it is considered that tungsten electrodes are scarcely worn because the melting point of tungsten is very higher than that of molybdenum. Thus, the tungsten electrodes were used. Figure 6 shows the optical micrographs of center part of the (Fe 0:6 Co 0:4 ) 72 Si 4 B 20 Nb 4 alloys. Molybdenum was used as the electrodes for the alloys shown in Figs. 6(a), (b). Tungsten was used as the electrodes for the alloys shown in Figs. 6(c), (d). The alternating electric current was fixed at 5 A, 5 khz. The magnetic fields were 0 T for the alloys shown in Figs. 6(a), (c) and 10 T for the alloys shown in Figs. 6(b), (d). The alloys produced at 0 T were cooled without the Table 2 Mo or W content by SEM-EDS corresponding to the optical micrographs shown in Fig. 6. Boron was excluded from detection elements because boron cannot be detected exactly by SEM-EDS. (a) (b) (c) (d) Mo (at%) W (at%) electromagnetic vibrations, and the alloys produced at 10 T were cooled under the electromagnetic vibrations. Table 2 shows Mo or W content by SEM-EDS corresponding to the optical micrographs shown in Fig. 6. Boron was excluded from detection elements because boron cannot be detected exactly by SEM-EDS. If molybdenum was used as the electrodes, Mo contents were 0.3 at% at 0 T and 0.8 at% at 10 T in the center of samples. However, if tungsten was used as the electrodes, there was no influence of electrodes in the center of samples. Thus, numbers of observed crystal particles and average areas of one particle in the micrographs shown in Fig. 6 were calculated. Results are shown in Table 3. The crystal particles with the area of more than 3 mm 2 were counted by computer. There was no difference between molybdenum and tungsten electrodes for average area of one particle. However, numbers of crystal particles

5 Effect of Electromagnetic Vibrations on Fe-Co-B-Si-Nb Bulk Metallic Glasses 57 Table 3 Numbers of observed crystal particles and average areas of one particle in the micrographs shown in Fig. 6. The crystal particles with the area of more than 3 mm 2 were counted by computer. Number of crystal particles with Mo electrodes (N/mm 2 ) Number of crystal particles with W electrodes (N/mm 2 ) Average area of one particle with Mo electrodes (mm 2 ) Average area of one particle with W electrodes (mm 2 ) were decreased when tungsten was used as electrodes. In particular, the crystal particles were observed hardly when the alloy was produced at 10 T, namely with the electromagnetic vibrations. However when molybdenum or tungsten was used as the electrodes, the effect of the electromagnetic vibrations was found to be the same, namely the electromagnetic vibrations act on decreasing the number of crystal nuclei. 4. Conclusions 0 T 10 T The effect of the electromagnetic vibrations on Fe-Co-B- Si-Nb bulk metallic glasses has been investigated, and the following conclusions have been derived. (1) The crystal particle were observed in a glassy phase. It was found that the crystal particle grew from one crystal nucleus. (2) Half round lines which consist of crystal particles in a glassy phase were observed in the boundary part of the molybdenum electrode for the alloy with the electromagnetic vibrations. It was considered that the reactants of the molybdenum electrode which were solid at 1573 K were moved into the sample by the electromagnetic vibrations and caused the nuclei of crystal particles which composed the half round lines. (3) If tungsten was used as the electrodes, there was no influence of electrodes in the center of the samples. When molybdenum or tungsten was used as the electrodes, the effect of the electromagnetic vibrations was found to be the same, namely the electromagnetic vibrations act on decreasing the number of crystal nuclei. Acknowledgments This work has been supported by a grant for the Metallic Glasses Project of the Material Industrial Competitiveness Strengthening Program from New Energy and Industrial Technology Development Organization (NEDO). The authors are grateful to Dr. N. Nishiyama, Mr. K. Amiya and Mr. A. Urata for providing the Fe-based alloys and numerous helpful discussions. The authors also thank Dr. K. Yasue, Mr. Y. Sakaguchi and Mr. H. Matsubara for technical assistance. REFERENCES 1) S. Asai: ISIJ Int. 29 (1989) ) C. Vives: Mater. Sci. Eng. A 173 (1993) ) C. Vives: Metall. Mater. Trans. B 27 (1996) ) A. Radjai, K. Miwa and T. Nishio: Metall. Mater. Trans. A 29 (1998) ) A. Radjai and K. Miwa: Metall. Mater. Trans. A 31 (2000) ) A. Radjai and K. Miwa: Metall. Mater. Trans. A 33 (2002) ) Y. Mizutani, S. Kawai, K. Miwa, K. Yasue, T. Tamura and Y. Sakaguchi: Mater. Trans. 45 (2004) ) Y. Mizutani, Y. Ohura, K. Miwa, K. Yasue, T. Tamura and Y. Sakaguchi: Mater. Trans. 45 (2004) ) Y. Mizutani, A. Kawada, K. Miwa, K. Yasue, T. Tamura and Y. Sakaguchi: Journal of Materials Research 19 (2004) ) T. Tamura, K. Amiya, R. S. Rachmat, Y. Mizutani and K. Miwa: Nature Materials 4 (2005) ) T. Tamura, R. S. Rachmat, Y. Mizutani and K. Miwa: Mater. Trans. 46 (2005) ) T. Tamura, D. Kamikihara, Y. Mizutani and K. Miwa: Mater. Trans. 47 (2006) ) K. Amiya, A. Urata, N. Nishiyama and A. Inoue: Mater. Trans. 45 (2004) ) A. Urata, K. Amiya and A. Inoue: ISMANAM 2004 Abstracts (2004) p. 142.